Hi,
HAProxy 3.5-dev12 was released on 2026/11/12. It added 126 new commits
after version 3.5-dev11.
About half of the patches are bug fixes (68), mostly for bugs we found
in resolvers, h2, h3, fcgi consecutive to our recent scans, though some
came from testing and reports as well. It's difficult to give examples
as they're spread all over the place, as usual. The rest, over all, is:
- a nice batch of QUIC harmless cleanups that will help maintain the
code in a more maintainable state;
- debugging: "haproxy -d" will no longer emit "connect"/"close" lines
for internal proxies (CLI and peers), as this was a real pain during
development when combined with external tools collecting data. It was
so easy that I regret not doing it 10 years ago.
- doc: some updates and clarifications to the threat model and the
security doc to try to improve the signal-to-noise ratio on the
security list(s). I'm adding an 's' because some even send messages
to haproxy.com, adding even more burden. Please don't do that.
- HTTP violations: the new "invalid-version-string" HTTP violation
checks that HTTP version advertised over an HTTP/1 connection is
exactly "1.0" or "1.1" which are the only two specified on such
connections. While haproxy itself is transparent to other versions,
other agents are having difficulties with them and can be confused.
This is enabled by default.
- config: for quite a while we've had discussions about how to warn
against accidental misuse of "http-request deny" combined with samples
that are subject to aliasing (e.g. the path). A reasonable solution
was implemented here. From now on, http-request actions that are used
to intercept traffic (deny, silent-drop, tarpit, reject, auth) check
at parsing time if they're combined positively with ACLs involving
some sample fetches subject to aliasing (path, query, url, ...). If
this is the case, a notice message will be emitted explaining what the
rule does and how to better handle these. For the cases where it's
intentional, such as blocking noisy bots etc, just adding "no-warn" on
the line before the if/unless keyword is sufficient to silence them,
and it serves as a config doc to mean "I know what I'm doing". Once
the config is clean, it can make sense to turn all of these into real
warnings via the dediated global keyword "on-unreliable-rules" in
order to easily catch any new accidental addition. If, on the
opposite, all of this is considered as noisy (e.g. migrating an old
config), "on-unreliable-rules" may also be used to silence them all at
once. An entire section was added in the config explaining how to
write safer rules and why deny+match is never safe. Let's see how that
fits, and please report feedback, it's not too late to even just change
the default if it ends up being too verbose.
- scheduler: Olivier noticed before 3.4 (but we forgot to handle it)
that the scheduler's maximum number of tasks (runqueue-depth)
continues to be derived from the number of threads, while each thread
has been running on its own scheduler instance for ages now. This
results in large machines using super small queue depth, causing a
high scheduling overhead for no benefit. The operation was removed and
the default depth of 280 tasks that applied to 1 thread is now valid
regardless of the number of threads. It's possible that some users with
large machines could notice a small drop in CPU usage.
- CLI: the "wait" and "show events" commands now support a new options,
"-u" making them uninterruptible. By "uninterruptible", I mean that
they won't stop if the standard input is closed, which typically
happens when commands are sent in non-interactive mode over socat,
typically in scripts. That was preventing "show events -w" from
waiting for new logs, and "wait srv-removable" from waiting for a
server to be removable. In order never to lose control of the CLI, the
number of concurrent uninterruptible commands is limited to 3, so that
aborted commands won't take all connections. A new "kill" command is
provided to kill pending uninterruptible commands ("kill unintr").
- on the http-hardening front, new sample fetch functions
{fs,bs}.hdng.{found,fixed} now return respectively the hardening
conditions that were found, and the ones which were fixed (delete or
reject for now). That's an integer, to be logged as a hex number for
example. The new "hdng_rej_name" converter returns the name of the
reject hardening rule present in the "fixed" sample. The goal here
is to know on a log line why something was rejected. In addition,
"debug counters" now also reports counters for hardening operations
so that it's possible to observe which ones match and how often. E.g:
$ socat - stats.sock <<< "debug counters cnt" | grep hard
21 CNT h1.c:514 h1_headers_to_hdr_list(): "hardening
HTTP_HRD_REJ_LCASE_METHOD condition met"
21 CNT h1.c:514 h1_headers_to_hdr_list(): "hardening
HTTP_HRD_REJ_LCASE_METHOD enforced"
38 CNT h1.c:860 h1_headers_to_hdr_list(): "hardening
HTTP_HRD_REJ_USCORE_HDR_NAM condition met"
38 CNT h1.c:1072 h1_headers_to_hdr_list(): "hardening
HTTP_HRD_DEL_USCORE_HDR_NAM condition met"
1 CNT h1.c:1370 h1_headers_to_hdr_list(): "hardening HTTP_HRD_REJ_CL_TE
condition met"
1 CNT h1.c:1370 h1_headers_to_hdr_list(): "hardening HTTP_HRD_REJ_CL_TE
enforced"
- checks: support for MySQL 8.0 SHA2 passwords was added for tcp-checks,
given that version 8.4 disables the old method.
- resolvers: the number of servers visited when looking for a record is
now limited to those waiting for the same record; previously all of
those after the first one were visited. It could have contributed to
some cases of excessive load and watchdog conditions in the past,
though there are still plenty of other reasons there for this.
- various CI and DOC updates, and new regtests (we have 422 passing now).
We really need more testing, the current feedback is still limited. I
know that everyone is busy at this period of the year, but even just
building and passing a config through "haproxy -c" to make sure it
continues to pass is helpful if you don't have the time to prepare a
deployment and watch it.
We still have a bunch of reports to analyse and process. They keep arriving
faster than we can process them. The good point is that most of them are
really for minor bugs (and just noise). The bad point is that some of the
low-importance ones that keep arriving take more time to handle because
shutting them once for all implies non-trivial changes.
I'm still aiming for a release at the end of the month, so that leaves
us about two weeks, thus more likely a single one for remaining changes.
We'll also have to plan new stable releases I think.
Please find the usual URLs below :
Site index : https://www.haproxy.org/
Documentation : https://docs.haproxy.org/
Wiki : https://github.com/haproxy/wiki/wiki
Discourse : https://discourse.haproxy.org/
Slack channel : https://slack.haproxy.org/
Issue tracker : https://github.com/haproxy/haproxy/issues
Q&A from devs : https://github.com/orgs/haproxy/discussions
Sources : https://www.haproxy.org/download/3.5/src/
Git repository : https://git.haproxy.org/git/haproxy.git/
Git Web browsing : https://git.haproxy.org/?p=haproxy.git
Changelog : https://www.haproxy.org/download/3.5/src/CHANGELOG
Dataplane API :
https://github.com/haproxytech/dataplaneapi/releases/latest
OpenTelemetry : https://github.com/haproxytech/haproxy-opentelemetry
Pending bugs : https://www.haproxy.org/l/pending-bugs
Reviewed bugs : https://www.haproxy.org/l/reviewed-bugs
Code reports : https://www.haproxy.org/l/code-reports
Latest builds : https://www.haproxy.org/l/dev-packages
Willy
---
Complete changelog :
Adam Crawford (1):
BUG/MEDIUM: filters: avoid NULL block access in decompression
Aleksandar Lazic (1):
BUG/MEDIUM: mux-fcgi: do not enforce the timeout on connections with
streams
Alex Joseph (1):
MINOR: tcpcheck: add post-80-sha2 mode to mysql-check for
caching_sha2_password
Alexandre Laroche (3):
BUG/MINOR: server: fix a leak when completing a named default-server
BUG/MINOR: ssl/ckch: fix an invalid free() in the domains/ips error paths
BUG/MINOR: tools: ha_freearray() must also free the array
Amaury Denoyelle (17):
MINOR: cli: factorize CLI session reset
MINOR: cli: store active CLI sessions in a global list
MINOR: cli: implement basic mechanism for uninterruptible jobs
MINOR: cli: implement uninterruptible mode for wait
MINOR: ring: support inactivity timeout in show ring CLI handler
MINOR: sink: add timeout argument for "show events"
MINOR: cli: add "kill" CLI command
REGTESTS: cli: check uninterruptible mode
BUG/MINOR: quic: reject invalid stream ID after MUX closure
BUG/MEDIUM: quic: limit reject after MUX closure to once per stream
BUG/MINOR: mux_quic: remove dangling qc->qcc ref on MUX init failure
MINOR: http: add <need_eot> to trailers parsing function
MINOR: mux_quic: refactor qcs_http_handle_standalone_fin()
BUG/MINOR: h3: fix missing EOM after trailers without FIN
BUG/MINOR: h3: close connection on response start-line alloc failure
BUG/MINOR: h3: fix error code on incomplete request
BUG/MEDIUM: h3: reject standalone FIN on invalid request/response
Christopher Faulet (31):
BUG/MEDIUM: stream: reset analyse_exp when analysers are stripped on error
BUG/MINOR: http-ana: also set FINAL on http-response actions on server
error
BUG/MEDIUM: spoe: stop waiting on a dead producer channel
BUG/MINOR: applet: accept a temporarily owned tid in appctx_init()
BUG/MEDIUM: mux-h2: consider the send list as pending data for the timeout
BUG/MEDIUM: mux-h2: refresh the connection timeout before parking it idle
BUG/MEDIUM: mux-fcgi: arm the connection timeout only when data remains
BUG/MEDIUM: mux-fcgi: detect an early connection error during the
handshake
BUG/MEDIUM: mux-fcgi: ignore conn->owner when deciding if a connection is
dead
MINOR: mux-fcgi: check for a dead connection before refreshing the timeout
MINOR: mux-fcgi: don't add a closed connection to an idle list on detach
BUG/MEDIUM: mux-h2: also consider the fctl_list as pending data for the
timeout
BUG/MINOR: mjson: fix out-of-bounds reads on malformed JSON paths
BUG/MEDIUM: resolvers: detach servers without resolvers from the SRV
request
BUG/MEDIUM: resolvers: don't ignore records not matching a resolve-net
family
BUG/MEDIUM: resolvers: skip the data of ignored additional records
BUG/MINOR: resolvers: reject a null or negative resolve_retries value
BUG/MINOR: resolvers: reject null "timeout resolve" and "timeout retry"
values
BUG/MINOR: resolvers: preserve nameserver identity on "clear counters all"
BUG/MINOR: resolvers: don't dump a nameserver twice in "show resolvers"
BUG/MINOR: resolvers: Always init a new resolution's list element
BUG/MINOR: resolvers: count nameserver timeouts
BUG/MEDIUM: resolvers: reject servers sharing an SRV record with
different resolvers
BUG/MINOR: resolvers: check the SRV expiration date again under the lock
OPTIM: resolvers: only walk duplicates when looking up an existing record
BUG/MINOR: resolvers: warn about ignored arguments on datagram nameservers
BUG/MINOR: resolvers: fix leaks on do-resolve's fetch availability error
BUG/MINOR: resolvers: report a timeout when a TCP fallback gets no
response
BUG/MINOR: resolvers: use truncated SRV responses when no TCP fallback is
possible
BUG/MEDIUM: resolvers: properly skip root-named authority and additional
records
BUG/MINOR: resolvers: compute a name's wire length by walking its labels
Frederic Lecaille (11):
CLEANUP: quic: compute the STREAM frame end offset once in
qc_stream_frm_is_acked()
CLEANUP: quic: fix and complete the comments in quic_retransmit.c
CLEANUP: quic: remove a redundant test in the fast retransmission lookups
CLEANUP: quic: use QUIC_PACKET_PN_MAXLEN instead of the magic 4
CLEANUP: quic: remove the dead encryption level variable in
qc_prep_hdshk_fast_retrans()
CLEANUP: quic: drop the intermediate packet tree pointer in the fast
retransmissions
CLEANUP: quic: remove the duplicate qc_prep_hdshk_fast_retrans() prototype
BUG/MINOR: fd: allow fd_write_frag_line() to run before fdtab allocation
BUG/MEDIUM: quic: fix loss time selection in quic_loss_pktns()
BUG/MINOR: quic: reject RETIRE_CONNECTION_ID for a CID not sent to the
peer
BUG/MINOR: stick-table: release server_key on any stksess destruction
Gihyeon Ryu (1):
BUG/MAJOR: quic: don't reference a retired CID from a NEW_CONNECTION_ID
frame
Lukas Tribus (1):
DOC: minor: update VTest2 URL everywhere
Martin Mirchev (2):
BUG/MINOR: tcpcheck: check the number of "hdr" in http-check send
BUG/MINOR: stick-table: reject invalid gpc/gpt array indexes
Olivier Houchard (3):
BUG/MEDIUM: ssl: don't use MSG_MORE for kTLS control records
BUILD: Makefile: force $(shell command -v ...) to run through the shell
BUG/MEDIUM: threads: don't read lock history entries that don't fit in
ulong
Owais Lone (1):
REGTESTS: protobuf: cover the ungrpc message realignment
Tiburce GIROUX (1):
BUG/MEDIUM: flt-decomp: preserve pending output before decoding again
William Dauchy (2):
BUG/MINOR: quic-be: fix Initial token length check on backend connections
BUG/MINOR: quic-be: drop long header packets with unsupported version
William Lallemand (10):
CI: github: switch development builds back to ubuntu-latest
CI: github: remove MIPS targets from cross-zoo.yml
CI: temporarily disable AWS-LC 3.x (fips-2024-09-27)
CI: github: install libcrypt-dev:i386 for the i686 build
BUG/MINOR: quic/ssl: record TLS version bounds for certificate selection
CI: github: pin ARM development builds to Ubuntu 26.04
BUG/MINOR: cfgparse: detect unsupported crypt() hashes on Debian/Ubuntu
DOC: configuration: clarify normalize-uri version availability
CI: github: re-enable AWS-LC 3.x (fips-2024-09-27)
BUG/MINOR: ssl: skip the cipher workaround with AWS-LC API 38 and newer
Willy Tarreau (40):
BUG/MINOR: http/hardening: don't count unregistered hdr/tlr too early
MINOR: http/hardening: turn it to a macro and count matches with
COUNT_IF()
MEDIUM: h1/violations: restrict valid version strings to "1.0" and "1.1"
BUILD: http: fix build of macro http_harden() with clang
MINOR: debug: add a COUNT() macro which doesn't take a condition
CLEANUP: debug/tree-wide: replace COUNT_IF(1, ...) with COUNT()
BUG/MEDIUM: compression: use a trash chunk as large as the buffer for the
output
BUILD: debug: fix _COUNT undefinied with USE_OBSOLETE_LINKER
CLEANUP: mux-h2: clarify that bptr is not NULL when calling to
h2_get_buf()
BUG/MEDIUM: mux-h2: always return already allocated buffers in
h2_get_buf()
BUG/MEDIUM: mux-h2: permit the buffer callback to allocate despite the
queue
BUG/MEDIUM: mux-h2: wake the connection up after a delayed mbuf allocation
BUG/MEDIUM: mux-h2: serve all pending buffer allocations in the buffer
callback
BUG/MINOR: mux-h2: unblock the demux when destroying the stream it was
blocked on
BUG/MEDIUM: mux-h2: don't send RST_STREAM for backend streams whose
HEADERS were not sent
BUG/MEDIUM: mux-h2: also emit the HPACK dynamic table size update on
requests
BUG/MINOR: mux-h2: also emit the HPACK Dynamic Table Size Update before
trailers
BUG/MINOR: mux-h2: only convert GET upgrade requests to Extended CONNECT
REGTEST: ocsp: make the ocsp_client_check test fast
BUG/MINOR: acl: check the presence of an argument to "-u"
BUG/MINOR: server: don't report dyncookie collision on same IP servers
SCRIPTS: create-release: do not match the word "version" isolated
MEDIUM: sched: fix the default runqueue-depth
DOC: security: clarify how to send proper reports
DOC: threat-model: add a few more entries for stuff that should be public
DOC: configuration: add a section on secure filtering of HTTP requests
MINOR: sample: add SMP_USE__MAY_ALIAS to mark fetches returning aliasable
contents
MINOR: http-fetch: mark URI-based sample fetches with SMP_USE__MAY_ALIAS
MINOR: acl: add acl_cond_find_positive_use() to find positive terms by
usage
MINOR: http-act: mark http-request access control actions with
KWF_ACCESS_CTRL
MINOR: http-rules: accept "no-warn" before the condition of access
control rules
REGTESTS: add a no-warn directive on deny rules using a path
MEDIUM: http-rules: report access control rules relying on aliasable
contents
MINOR: http: add a global setting to decide on warnings for http rules
DEBUG: stream: do not emit debug messages for internal proxies
MINOR: mux-quic: add an sctl() callback to the application ops
MINOR: mux: add sctl MUX_SCTL_GET_HDNG_FOUND/FIXED to report hardening
results
MINOR: http-fetch: add {fs,bs}.hdng.{found,fixed} to report hardening
results
MINOR: http-conv: add hdng_rej_name to name the HTTP hardening reject rule
REGTESTS: http-messaging: test the reporting of HTTP hardening results
---